JP2002319924A - Method and device for processing data signal - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method and a device for transmitting/receiving a digital image by using an electronic image watermark for decoding. SOLUTION: In order to decode a received digital signal with an electronic watermark by means of the known electronic watermark, by using a parameterisable iterative decoder, which can be made into parameter, at least one part of the digital signal is decoded, the electronic watermark is extracted from the decoded signal and the extracted electronic watermark is compared with the known electronic watermark. According to the compared result, at least one parameter of the decoder is corrected as needed.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a method and apparatus for processing digital signals, and more particularly to a method and apparatus for transmitting and receiving digital images using an image digital watermark for decoding. is there.

[0002] More precisely, the invention relates to the transmission of a digital image in which a known watermark is inserted on a noisy channel, wherein the system for the protection against transmission errors comprises a parameterizable iterative decoding. I.e., using soft-decision decoding, which can change a parameter indicating, for example, the number of iterations.

[0003]

BACKGROUND OF THE INVENTION As a non-limiting example, the present invention is described herein in terms of its application to turbo code.

[0004] Turbo codes have a signal-to-noise ratio (SNR).
Is highly advantageous in conditions that are low, more specifically those encountered on wireless links where there is a large distance between the transmitter and the receiver.

A conventional turbo encoder comprises two recursive systematic convolutional (RSC) encoders arranged as shown in FIG. 1 and an interleaver. The turbo coder outputs its triples of binary elements (x, y1, y2), where x is the so-called systematic output of the turbo coder, i.e., which has not undergone any processing compared to the input signal x. Where y 1 is the output encoded by the first RSC encoder and y 2 is the second RS after passing through the interleaver.
This is the output encoded by the C encoder.

For more information on turbo codes, see C. B.
errou, A. Glavieux, P. Thitimajshima, Near Shann
on Limit error-correcting coding and decoding: tu
You can refer to the paper entitled "rbo-codes"(ICC'93, Geneva), which is valid.

FIG. 2 shows an example of a conventional turbo decoder capable of decoding data supplied by a turbo encoder as shown in FIG. The inputs x, y1, y2 of the turbo decoder are the outputs of the turbo encoder that are received by the decoder after passing through a transmission channel considered herein to be ideal for ease of description. The structure of such a turbo decoder is well known to the expert and therefore will not be described in detail herein.

As can be seen from FIG. 2, the complete decoder is
It has a very complicated structure. In particular, two decoders, shown in FIG. 2 as “decoder 1” and “decoder 2”,
For example, BCJR type, ie Bahl, Cocke, Jelinek, Ra
One that uses the algorithm of viv or one of the SOVA (soft output Viterbi algorithm) type is required.

[0009] The conventional turbo decoder also needs to feed back the output of the deinterleaver π2 to the input of the first decoder in order to transmit so-called “foreign” information from the second decoder to the first decoder. is there.

[0010] A signal protection system using a turbo code is as follows.
In order to decode a received signal, it is necessary to perform a certain number of decoding iterations during reception.

The problem is to know how many iterations to take on reception to guarantee good quality, the goal is to save computation time while maintaining good quality of the decoded data, And minimizing the number of iterations to gain speed. Therefore, it is advantageous to flexibly adapt the number of iterations to decoding.

On the other hand, it is possible to insert supplementary information into the data using a digital watermarking system.

The digital watermarking of an image consists of directly inserting a non-erasable digital watermark into the digitized data, which assimilates the encoding of supplementary information in the data. .

General state-of-the-art watermarking for fixed images and video is described in M. Swanson, M. Kobayashi, A.
Multimedia data-embedding and watermark by Tewfik
ingtechnologies ”(Proc. of the IEEE, Vol. 86, No.
6, 1998).

It is preferred to use so-called invisible digital watermarks which exhibit the following factors:

This watermark is not perceptible. That is, even if such a digital watermark is inserted, the perceptual quality of digital data must be maintained. The inability to perceive the watermark also complicates piracy.

The digital watermark cannot be erased. That is, in order to counter a deliberate attack for destroying the digital watermark, the digital watermark cannot be statistically detected in the digitalized digital data.

There are several classes of digital watermarks for different purposes.

The first category includes copyright protection methods, for which the watermark must be robust against the various types of processing applied to the image: compression, reformatting, and filtering.

The second category relates to methods intended to authenticate the digital data itself and to reveal any possible changes to the image. In this second case, the fragile method will be described.

An intermediate class of methods is the so-called semi-fragile method, which makes it possible to obtain strength for a specific fixed distortion (for example, compression with a given compression ratio), This type of image correction cannot be obtained.

Up to the present, the turbo code used for an image into which a digital watermark is to be inserted aims at protecting only the digital watermark itself, but not the image. Regarding this point, the International Conference on Image Processing (International Conferen
S. Perei, published in ce on Image Processing)
ra, S. Vooshynovski and T. Pun, “Effective chann
el coding for DCT watermarks ”is effective.

It is further known that watermarks inserted into images can serve to estimate the quality of the channel. This was discussed in September 2000 by the International Symposium on Turbo-codes & French Symposium on Turbo-codes and related topics.
Relative Topics), EA Ratzer and
Codes for channels with inserti by DJC Mackay
The article entitled "Ons, deletions and substitutions" can be referred to and is valid.

In the case of turbo codes, the number of iterations for decoding is generally fixed empirically or according to a predetermined stopping criterion. For example, US Pat. No. 5,761,248 describes a stopping criterion for iterative decoder iterations based on entropy.

Another method of fixing the number of repetitions consists of transmitting a known sequence at the start before transmitting the data. Decoding the pilot sequence makes it possible to estimate the number of iterations to be performed. This technique has the disadvantage that the number of iterations can only be estimated once, rather than continuously.

[0026]

SUMMARY OF THE INVENTION It is an object of the present invention to use a known watermark from a receiver as a pilot in the data to be transmitted and to decode the watermark before decoding the quality of the transmission channel. It is to correct the aforementioned drawbacks by obtaining an estimate and thus deducing the number of iterations required to obtain a given reception quality.

[0027]

To this end, the present invention proposes a method for transmitting digital signals, the proposal comprising the steps of:-obtaining a digital signal known to the receiving end so as to obtain a digital signal with a digital watermark; Performing a watermarking operation that inserts the digital watermark into the digital signal, and encoding the digital signal with the watermark by the encoder so as to obtain a coded watermark that can be decoded by the iterative decoder with adjustable parameters. Performing an encoding operation;-transmitting the encoded watermarked signal, whereby at the receiving end at least one parameter of the decoder is obtained from the encoded watermarked signal; Performing a transmission operation that can be modified according to the result of the comparison between the data and the known digital watermark.

The invention thus makes it possible to use the watermark inserted in the picture as a pilot on reception and to determine the number of iterations required to decode the picture with satisfactory quality.

Given that the data and the watermark simultaneously pass through the transmission channel, they both degrade in the same way.
Thus, the channel estimation performed by the digital watermark is thus valid for this data.

In this way, a stopping criterion is obtained which allows to know how many iterations need to be performed.

Further, the number of repetitions to be performed over time is continuously updated.

In one specific embodiment, the encoder is a turbo encoder.

[0033] Turbo codes provide good protection of data against transmission errors, and the number of uncorrected errors decreases very sharply instead of a small increase in the signal to noise ratio.

In one specific embodiment, a modulation operation for modulating the encoded watermarked signal is also performed before the transmission operation.

The resulting modulated signal has characteristics well suited to the required use conditions. For example, modulation shifts the spectrum of a signal to other frequency domains without loss of information to meet transmission / reception constraints (antenna efficiency and size) and the conditions imposed by the transmission channel (propagation, usable Bandwidth) or facilitate certain signal processing operations.

Modulation also allows the realization of sharing of the communication channel between several signals transmitted simultaneously (frequency multiplexing, assigning a different frequency band to each message transmitted simultaneously, sampled for each message) Time multiplexing to transmit values sequentially).

Modulation also allows the amplification of weak low frequency signals and the achievement of effective filtering, especially by being relieved of 1 / f background noise.

Modulation may also include, for example: recording a signal whose spectrum extends to zero frequency on a magnetic medium (measurement recorder); conditions for detection (radar) and immunity from noise (angylar modulation and pulse). (Modulation) to improve the spectrum of the signal to be transmitted, to make communication more secretive and to make jams more difficult (spread spectrum systems), automatic adjustment, or an appropriate amount of control of a mechanical or industrial process. It is possible to change.

According to one particular property, the modulation operation is based on the QP
It consists of the modulation of an encoded digital watermarked signal by SK type modulation.

Phase modulation does not raise the power of the signal to be transmitted, but rather modulates the phase or frequency of the sine carrier in proportion to the signal that modulates the phase or frequency of the sine carrier. At the cost (which is a non-linear operation), it has the advantage of increased immunity to noise.

As a variant, the modulation operation consists in modulating the coded digital watermarked signal by means of an OFDM type modulation.

OFDM modulation offers the advantage of distributing the signal to be transmitted in parallel over a number of individually modulated subcarriers at a fraction of the rate of the data to be transmitted. Since the rate is low for each individual subcarrier and the associated bandwidth is less than the overall bandwidth, there is less risk that the amplitude and phase characteristics will vary over the spectrum of frequencies that make up each individual bandwidth. Therefore, it is possible to obtain good transmission quality at a high rate, but at the cost of obviously using more resources.

According to one particular characteristic, the watermarking operation uses a fragile or semi-fragile type technique.

This makes it possible to characterize the loss of information due to transmission.

In one particular embodiment, the digital signal is an image signal.

The area of image processing actually constitutes an advantageous field of application of the invention.

For the same purpose as described above, the present invention also provides
A transmission device for transmitting a digital signal, the proposal comprising: a digital watermarking unit for inserting a digital watermark known to a receiving side into the digital signal and outputting a digital signal with a digital watermark; An encoding unit for encoding the digital watermarked signal by an encoder and outputting an encoded digital watermarked signal that can be decoded by a parameter adjustable iterative decoder; and transmitting the encoded digital watermarked signal. A transmission module for enabling the receiver to modify at least one parameter of the decoder according to a comparison between a watermark obtained from the encoded watermarked signal and a known watermark And characterized in that:

The specific characteristics and advantages of the transmitting device are the same as those of the transmitting method and will not be repeated here.

For the same purpose, the present invention also provides
A decoding method is proposed for decoding a received digital signal, i.e. a digital signal that has been watermarked with a known watermark, the proposal comprising: decoding at least a part of the digital signal with a parameter-adjustable iterative decoder; Performing a decoding operation; performing a digital watermark extraction operation for extracting a digital watermark from the decoded signal; performing a comparison operation for comparing the extracted digital watermark with a known digital watermark. Performing a correction operation to correct at least one decoding parameter if necessary according to the result of the comparison.

Thus, it is possible to use a known watermark inserted in the data as a pilot and parameterize it to improve decoding.

According to one particular characteristic, the modifying operation comprises:-determining the number of iterations to be applied as a parameter of decoding according to the result of the comparison; calculating the number of iterations to be applied to decoding; Adjusting the decoding by applying the determined number of repetitions during decoding.

Thus, the invention makes it possible to determine, by means of a known watermark inserted in the data, the most suitable number of iterations for decoding.

In one particular embodiment, the decoding operation decodes the data with an iterative decoder by applying repetition or half-iteration to obtain a partially decoded digital signal with a watermark. A partial decoding operation, wherein the correction operation comprises:-a quality test operation consisting of testing whether the quality of the extracted digital watermark is satisfactory; and-a decoding parameter unless the quality is satisfactory. Include additional iterations or half-iterations during decoding so that the optimal number of iterations or half-iterations to apply is finally obtained.

This allows only the exact number of iterations or half-iterations required, which is a gradual adjustment over time.

In one particular embodiment, the iterative decoder is a turbo decoder.

This particular embodiment has the same advantages as using a turbo coder during transmission.

In a preferred application of the invention, the digital signal is an image signal.

For the same purpose, the present invention also provides
Proposing a receiving method for receiving a digital signal, the proposal comprising: performing a receiving operation for receiving modulated symbols; and receiving received modulated symbols so as to obtain a demodulated signal. Performing a demodulation operation of demodulating the data; and performing a decoding operation of decoding data demodulated using the above-described decoding method.

The demodulation operation is performed by QPSK type modulation or OPSK modulation.
It can be configured by applying demodulation corresponding to FDM type modulation.

For the same purpose, the present invention also provides
A decoding device for decoding a received digital signal, i.e. a digital signal watermarked with a known watermark, is proposed, comprising: decoding at least a part of the digital signal with a parameter-adjustable iterative decoder; A watermark extraction unit for extracting a watermark from the decoded signal; a comparison unit for comparing the extracted watermark with a known watermark; and according to the result of the comparison. , A modification unit for modifying at least one decoding parameter.

The specific characteristics and advantages of the decoding device are the same as those of the decoding method and will not be repeated here.

For the same purpose, the present invention also provides
Proposes a receiving device for receiving a digital signal, the proposal comprising: a receiving unit for receiving modulated symbols; and a demodulation for demodulating received modulated symbols and outputting demodulated data. And a decoding unit for decoding data demodulated by the above-described decoding device.

The present invention also provides the transmission method and / or
Or a digital signal processing device having means adapted to implement a decoding method.

The present invention also relates to a transmitting apparatus and / or
Alternatively, the present invention relates to a digital signal processing device having a decoding device.

The present invention also provides the transmission method and / or
Or a telecommunications network having means adapted to implement a receiving method.

The present invention also provides a transmitting apparatus and / or
Or a telecommunications network having a receiving device.

The present invention also provides the transmission method and / or
Or a mobile station in a telecommunications network having means adapted to implement a receiving method.

The present invention also provides a transmitting apparatus and / or
Or a mobile station in a telecommunications network having a receiving device.

The present invention also provides the transmission method and / or
Or a base station in a telecommunications network having means adapted to implement a receiving method.

The present invention also provides a transmitting device and / or
Or a base station in a telecommunications network having a receiving device.

The invention also provides: an information storage means readable by a computer or a microprocessor which stores the instructions of the computer program and which makes it possible to carry out the transmission and / or decoding methods described above; Partially or wholly removable information storage means readable by a computer or microprocessor storing the instructions of the program and making it possible to carry out the transmission and / or decoding methods described above.

The present invention also provides the transmission method and / or
Or a computer program product comprising a sequence of instructions for performing a decoding method.

The specific characteristics and advantages of the various digital signal processing devices, the various telecommunication networks, the various mobile stations, the information storage means, the receiving method and the device of the computer program product are the same as those of the transmitting method according to the invention. Therefore, these particular characteristics and advantages will not be repeated here.

Other aspects and advantages of the present invention will become apparent from the following detailed description of particular embodiments of the invention, given by way of non-limiting example. In the description, reference is made to the accompanying drawings.

[0075]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS As shown in FIG. 3, a transmitting apparatus according to the present invention receives source data at an input unit, and inserts a certain watermark or pilot known in a receiver into the source data. It has a digital watermark insertion module 31 for generating a corresponding stream of data.

The digital watermarking method used in the present invention can be used to characterize information loss due to transmission.
It is of the fragile or semi-fragile type.

The data stream generated by module 31 is sent as an input to turbo coder 33 before being transmitted by radio module 35 and antenna 37 over a transmission channel.

A digital watermark is an additional information item, and is made up of one or more information bits arranged in a predetermined order.
This information is imperceptibly inserted into the source data. 2. Description of the Related Art Conventionally, a digital watermarking system is sometimes called an encoder, and includes two parts: a part for inserting a digital watermark, and a part for extracting a digital watermark, also called a decoder.

FIG. 4 shows an overall functional diagram of digital watermarking.
I is the original image, I 'is the image with digital watermark, K is the secret key, and W is the inserted information.

After the processing (compression, transmission, etc.), the modified image I ′ of the digital watermarked image I ′
Is obtained at the decoder. A key K 'is required to extract a digital watermark. In many methods, K '= K.
An estimated version of the digital watermark W represented by W 'is extracted. Note that the key is used in cases where it is desired to make digital watermark piracy more difficult. This parameter is optional if no protection is needed,
Not used here.

In the case of the present invention, the inserted information W is completely known to the decoder, and uses the comparison of W and W ′ to evaluate the quality of the received image, as will be shown later. It is assumed that a step of estimating a digital watermark, which is intended to be performed, is added.

Next, by way of non-limiting examples, two digital watermarking techniques that can be applied in the context of the present invention will be described.

According to the first digital watermarking technique shown in FIG. 5, inserting a digital watermark divides an image into blocks (step 50), and applies a discrete cosine transform (DC) to each block.
T) is applied to the set of spatial frequency coefficients obtained by applying (step 52).

This technique is used in particular in the JPEG compression standard with blocks of size 8 × 8. W is a binary image thumbnail (logo, a set of pseudo-random values) of size (M / 8) × (N / 8) when the image I is of size M × N.

After applying the DCT, a digital watermark W is inserted (step 54). Note that step 52 is followed by a coefficient-by-coefficient scalar quantization step if one wishes to perform watermarking compatible with the compression of the digital image. Next, the digital watermark is inserted with the quantized / dequantized coefficients.

For each DCT block, a set of coefficients whose absolute value is greater than a predetermined threshold is pre-selected. The bits of the digital watermark W are associated with each DCT block by spatial matching. That is, if the exponent is associated with a video scanning sequence, bit Wk is the kth DC
Inserted in the T block.

The original insertion is performed in two steps. For each selected coefficient, a check bit is calculated from the most significant bit representing the value of the coefficient. For example, if the coefficient is coded with 12 bits, the parity of the sum of the 10 most significant bits is taken.

Then, the least significant bit (LSB) is
Exclusive O of Wk and check bit based on current coefficient
Replace with the result of the R operation (represented by XOR).

This operation is repeated for all the coefficients selected in each block to be processed.

Next, during step 56, an inverse discrete cosine transform is performed on each block. Next, an image I ′ with a digital watermark is obtained.

At the decoder, the steps shown in FIG. 6 are applied.

The step 60 for dividing into blocks and the DCT 62 for each block are applied again. The same criteria as for the pre-selection of the coefficients that are assumed to have been watermarked during encoding also apply.

Next, in step 63, the original extraction of the digital watermark is performed. For each coefficient selected in each DCT block, the exact same formula as in encoding, ie,
Recalculate the check bits, for example, by applying the parity of the sum of the most significant bits.

The digital watermark estimate w ＾ 'k is equal to the result of an exclusive OR operation of the least significant bit of the processed coefficient and the check bit.

In order to evaluate the extracted watermark W ′ by comparison with a known watermark W, it is possible to calculate the proportion of the extracted erroneous binary values W′k.

For each DCT block, it is possible to have several extraction values according to the number of preselected coefficients in that block.

The true value Wk is also known. If one wishes to obtain a very high quality reproduction, the inserted bits Wk
Can be determined to be erroneous as soon as one of the corresponding extracted values is erroneous.

Alternatively, if it is desired to obtain an average quality, it is possible to first select a large number and, if a large number of extracted values are erroneous, estimate that the result is erroneous.
Therefore, for each bit of W, a diagnosis is obtained according to whether the extracted value is correct or erroneous.

The evaluation of the quality of the extracted digital watermark W ′ with respect to the known digital watermark W can be constituted only by calculating the ratio of the extracted value of the error. 50%
In these cases, it is necessary to add iterations to the decoder to improve the quality of the decoding.

It should be noted that the quality of the extracted watermark W ′ can be evaluated spatially. Then, if necessary, it is possible to determine that the decoding is to be repeated only with the subparts of the data that were not particularly well decoded.

Next, a description will be given of a second digital watermarking technique that can be used, which is disclosed in EP-A-1043687.

This is a "spread spectrum" type method in which a set of image coefficients is modified by adding a pseudo-random signal to insert each information bit. The advantage of this method is that the decoding of the watermark is performed by means of a statistical test which makes it possible to correlate the probability of error immediately.

For example, an image (1
Suppose a single information bit is inserted.
It can be easily extended to the case where some information bits are to be inserted.

The principle of digital watermark insertion / extraction is as follows. Consider a set of N coefficients representing an image, i.e., represented by Xi (1≤i≤N) and read in a predetermined order. A known law (eg, a uniform rule of [-1, 1])
Of the N pseudo random numbers, that is, ni, are extracted, and the following equation is applied.

X′i = Xi + αi · ni 1 ≦ i ≦ N (1) According to the consideration regarding the visibility of the inserted digital watermark,
Calculate the weight coefficient αi.

Decoding is performed by testing for a correlation between a set of tested coefficients X'i and the same pseudo-random signal as the encoder. Let Yi = X'i.ni. If the digital watermark insertion was actually performed as described above, Y = (Y
1,..., Yi,..., YN) must tend toward E (αi · ni2). Here, E represents a mathematical expected value. However, since the weight is positive, this value can be calculated and is a distance from zero. On the other hand, if no digital watermark has been inserted, the average of Y must tend to go to zero.

"Probability and statistic" by Papoulis
The statistical test described in the book entitled "s" (Prentice Hall) is applied to the mean of Y.

[0108]

T = E [Y] · √N / σ Y (2) where σ is the standard deviation of Y.

According to the standardized Gaussian rule, the test values are concentrated around 0 when no digital watermark is inserted (assumed H0), and when the digital watermark is inserted, the non-zero value m is obtained. Concentrate around (assumed H1). Thus, it is possible to characterize the probability of error in relation to the assumption H0 and to determine the detection limit threshold Td.

This method is now applied in the following manner.
In the encoder, a recursive search is performed on the smallest sized image blocks for which the detection criteria are to be validated (as described in EP 1043687). The detection criterion is t ≧ Ti, where Ti is a predetermined value. Advantageously, Ti is chosen to be large, for example equal to 10 or 15. Insertion is applied as shown in equation (1) above.

In the encoder, EP 104368
As shown in No. 7, a search algorithm is applied to a block into which a digital watermark is inserted, and the block has a decoding threshold Td such that Td <Ti. For example, Td = 4.

At the time of decoding, a detection test is calculated for the received data as shown in equation (2) above.

The real value of the variable t is an excellent estimate of the extraction quality of the digital watermark.

If there are no blocks of the image for which Td ≦ t ≦ Ti, it can be immediately deduced therefrom that the quality of the decoding is insufficient.

If the insertion is found to be corroborated, it is possible to make a decision on the quality of the extraction in relation to the position of t in the interval [Td, Ti] and consequently decide on an additional iteration. It is possible to

For example, the standardized distance between t and Ti can be related to a quantification of the quality of the data.
For example, if the distance is between 0 and 10%, no further iteration is performed during turbo decoding, and if it is between 10% and 20%, an additional iteration is performed.

As shown in FIG. 7, at the time of reception, unknown data received by the antenna 71 and the radio frequency unit 73 is first decoded by the turbo decoder at a certain number of repetitions, for example, four times. This example is in no way limiting. The watermark is then extracted from the data and compared to a known non-transmitted version in unit 77 to extract and compare the watermark.
According to the result of this comparison, the number of iterations of the turbo decoder can be modified and the quality of the channel can be taken into account.

FIG. 8 shows a transmission method according to the present invention in a preferred embodiment.

During a first step 80, a watermark is inserted into the data to be transmitted using a watermarking encoder according to one of the techniques described above.

Next, during step 82, the watermarked data is encoded by the turbo encoder.

Next, during step 84, step 8
The data obtained at the end of step 2 is modulated, for example, by QPSK modulation or OFDM modulation.

In this way, modulated symbols are obtained and transmitted during step 86.

The receiving method according to the present invention is shown in FIG. 9 in a preferred embodiment.

[0124] The first step 90 consists in receiving the modulated data transmitted by the transmitter, for example QPSK symbols or OFDM symbols.

Next, these symbols are set in step 92.
Demodulated during

Next, step 94 consists of turbo-decoding the received and demodulated data, for example, with four iterations. This example is in no way limiting.

During step 96, the watermark is extracted and compared to the full version known to the receiver.

Then, during step 98, the number of iterations actually required is calculated according to the digital watermark estimation performed in the previous step.

Finally, during step 99, step 9
The turbo decoding of the data is performed with the number of repetitions calculated in step 8.

FIG. 10 shows a modification of the receiving method of the present invention.

In this modification, the receiving step 100 and the demodulating step 102 are the same as steps 90 and 92 in the embodiment of FIG. 9, respectively.

Next, step 104 consists of turbo decoding the received and demodulated data in a first or half iteration. Here, a complete repetition corresponds to data passing through two decoders of a turbo decoder (see FIG. 2), while a half repetition corresponds to data passing only one decoder. It is worth mentioning that it will correspond.

During each decoding iteration and even half an iteration, the watermark is extracted at step 106 and compared to the full version known to the receiver.

During test 108, it is tested whether the desired quality has been achieved;
Subsequent iterations through zero or full or half repetitions.

If a satisfactory quality is obtained, decoding stops and the number of iterations actually performed is stored. The turbo-decoded data for this number of iterations or half-iterations is then delivered as output during step 112.

As shown in FIG. 11, the network according to the present invention is known as a base station SB, a station indicated by reference numeral 64, and reference numerals 661 and 66, respectively.
Some peripheral stations SP indicated by 2, ..., 66M
i, i = 1,..., M (M is an integer of 1 or more). The peripheral stations 661, 662, ..., 66M are the base stations S
B, each being connected by a radio link to a base station SB and able to move relative to the latter.

FIG. 12 is a diagram showing the peripheral stations SPi, i = 1,..., M according to the present invention in more detail, and includes the data source 20 and the receiving device 22.

The peripheral station SPi includes, for example, a digital camera, a computer, a printer, a server, a facsimile machine, a scanner, or a digital photographic device.

The transmission device 22 includes a calculation unit CPU (English “Central Processing Unit”) 26, a temporary data storage means 28 (RAM memory), and a data storage means 30
(ROM memory), for example, a character input means 32 such as a keyboard, an image reproducing means 34 such as a screen, and a data processing unit 24 including an input / output means 36.

The RAM 28 stores, in various registers: input data “data1_i” coming from the data source 20;
n "output data" data1_out "obtained at the end of the transmission method of the present invention, and-the current state of the series of bits coming from the turbo encoder in one particular embodiment in which the turbo encoder has two parities. It contains elements x, y1, y2 and a watermark W to be inserted into the data.

The peripheral station SPi also includes a transmitting unit 38 and a radio module 40 with a known transmitter with one or more modulators, filters and radio antennas (not shown).

The transmitting device 22 is stored in the ROM 30,
The steps of the transmission method shown in FIG. 8 can be executed by the program “Pe” whose instruction sequence corresponds to the steps of the transmission method of the present invention.

The peripheral station SPi according to the present invention also comprises, as shown in FIG. 13, a receiving device 70, a receiving unit 72, and an antenna thereof comprising a data processing unit corresponding to the data processing unit 24 already described with reference to FIG. And a wireless module 40 shown in FIG.

In the receiving device 70, the RAM 28 stores in various registers:-the input data "data2_in" coming from the transmitter after passing through the transmission channel;-the receiving method in one of the embodiments of the present invention Output data "data2_out" obtained at the end of the sequence, and-an estimate x ＾, y ＾ 1, y ＾ corresponding to the current element x, y1, y2 of that series of bits coming from the turbo encoder used by the transmitter. Including 2.

The receiving apparatus 70 stores the instruction sequence stored in the ROM 30 using the digital watermark W stored in the ROM 30 and the instruction sequence corresponding to the steps of the receiving method according to the present invention in the two embodiments.
The steps of the embodiment shown in FIG. 9 or the steps of the embodiment shown in FIG. 10 can be executed by “1” and “P2”.

[Brief description of the drawings]

FIG. 1 is a schematic diagram showing a structure of a conventional turbo encoder.

FIG. 2 is a schematic diagram showing a structure of a conventional turbo decoder.

FIG. 3 is a schematic diagram illustrating a transmission device according to a specific embodiment of the present invention.

FIG. 4 is a schematic diagram showing functions of a conventional image digital watermarking system.

FIG. 5 is a schematic diagram showing an image digital watermarking technique that can be used in the present invention.

FIG. 6 is a schematic diagram illustrating a technique for extracting a digital watermark from an image that has been digitally watermarked by a technique such as that shown in FIG.

FIG. 7 is a schematic diagram illustrating a receiving device according to a specific embodiment of the present invention.

FIG. 8 is a flowchart showing main steps of a transmission method according to a specific embodiment of the present invention;

FIG. 9 is a flowchart showing main steps of a receiving method according to a specific embodiment of the present invention;

FIG. 10 is a flowchart showing main steps of a receiving method according to another embodiment of the present invention.

FIG. 11 is a schematic diagram illustrating a telecommunications network according to a particular embodiment of the present invention.

FIG. 12 is a schematic diagram illustrating a mobile station in the telecommunications network shown in FIG. 11 used for transmission according to one particular embodiment of the present invention.

FIG. 13 is a schematic diagram illustrating a mobile station in the telecommunications network shown in FIG. 11 used for reception according to one particular embodiment of the present invention.

[Explanation of symbols]

 Reference Signs List 20 data source 22 transmission device 24 data processing unit 26 CPU 28 temporary data storage means 30 data storage means 31 digital watermark insertion module 32 character input means 33 turbo encoder 34 image reproduction means 35 wireless module 36 input / output means 37 antenna 38 transmission Unit 40 Wireless module 64 Base station 66 Peripheral station 70 Receiving device 71 Antenna 72 Receiving unit 73 Radio frequency unit 77 Unit

 ──────────────────────────────────────────────────の Continued on the front page (72) Inventor Frederick Hermann Rennes-Attarant France, Cedex Cesson-Sevigne 35517, Rue de la Touch-Lambert Canon Research Center France S.S. A. (72) Inventor Ioana Donescue, Rennes-Attarant, France Cedex Cesson-Sevigne 35517, Rue de la Touch-Lambert Canon Research Center France S.S. A. F term (reference) 5B057 CA12 CA16 CB12 CB18 CB19 CE08 CG02 5C076 AA14 BA07 BA09 5J065 AA01 AB02 AC02 AD10 AE01 AE06 AF02 AG06 AH15 AH20 5K014 AA01 BA10 FA16 GA03 HA06

Claims (52)

[Claims] 1. A transmission method for transmitting a digital signal, comprising: performing a digital watermarking operation of inserting a digital watermark known to a receiving side into a digital signal so as to obtain a digital signal with a digital watermark. Performing an encoding operation to encode a digital signal with a watermark by an encoder so as to obtain an encoded digital watermark that can be decoded by an iterative decoder with adjustable parameters; and By transmitting the watermarked signal, the receiving side can modify at least one parameter of the decoder according to a comparison between a watermark obtained from the encoded watermarked signal and a known watermark. Performing an operation. 2. The transmission method according to claim 1, wherein the encoder is a turbo encoder. 3. The transmission method according to claim 1, further comprising, before the transmission operation, a modulation operation of modulating the encoded digital watermarked signal. 4. The transmission method according to claim 3, wherein the modulation operation modulates the encoded signal with a digital watermark by QPSK type modulation. 5. The transmission method according to claim 3, wherein the modulation operation modulates the encoded digital watermarked signal by OFDM type modulation. 6. The transmission method according to claim 1, wherein the digital watermarking operation uses a fragile or semi-fragile type technique. 7. The transmission method according to claim 1, wherein the digital signal is an image signal. 8. A transmitting device for transmitting a digital signal, comprising: a digital watermarking means for inserting a digital watermark known to a receiving side into the digital signal and outputting a digital signal with a digital watermark; Encoding means for encoding a digital signal with a watermark by a device and outputting an encoded digital watermarked signal which can be decoded by a parameter adjustable iterative decoder; and transmitting the encoded digital watermarked signal, At the receiving end, at least one parameter of the decoder is
A transmitting unit configured to correct the digital watermark according to a comparison result between a digital watermark obtained from an encoded digital watermarked signal and a known digital watermark. 9. The transmission device according to claim 8, wherein the encoder is a turbo encoder. 10. The transmitting apparatus according to claim 8, further comprising a modulating unit for modulating the encoded signal with a digital watermark. 11. The transmitting apparatus according to claim 10, wherein the modulating means modulates the encoded signal with a digital watermark by QPSK type modulation. 12. The transmitting apparatus according to claim 10, wherein the modulating means modulates the encoded digital watermarked signal by OFDM type modulation. 13. The transmitting apparatus according to claim 8, wherein said digital watermarking means uses a fragile or semi-fragile type technique. 14. The transmitting device according to claim 8, wherein the digital signal is an image signal. 15. A decoding method for decoding a received digital signal that has been digitally watermarked with a known digital watermark, the decoding operation comprising decoding at least a portion of the digital signal with a parameter adjustable iterative decoder. Performing, performing a digital watermark extraction operation for extracting a digital watermark from the decoded signal, performing a comparison operation for comparing the extracted digital watermark with a known digital watermark, and according to a result of the comparison. Performing a correction operation to correct at least one decoding parameter when necessary. 16. The method according to claim 1, wherein the correcting operation includes: determining a number of repetitions to be applied as a parameter of decoding according to a result of the comparison; calculating a number of repetitions to be applied to decoding; Adjusting the decoding applied during the decoding. 17. The decoding operation is a partial decoding operation in which repetition or half-repetition is applied and data is decoded by an iterative decoder so as to obtain a partially decoded digital signal with a watermark. The modifying operation comprises: a quality test step consisting of testing whether the quality of the extracted digital watermark is satisfactory; and, unless the quality is satisfactory, an optimal iteration or half-iteration to be applied as a parameter for decoding. 16. The decoding method according to claim 15, comprising: performing an additional iteration or half-iteration during decoding so that the number is obtained last. 18. The decoding method according to claim 15, wherein the iterative decoder is a turbo decoder. 19. The decoding method according to claim 15, wherein the digital signal is an image signal. 20. A receiving method for receiving a digital signal, comprising: performing a receiving operation of receiving a modulated symbol; and converting the received modulated symbol to obtain a demodulated signal. A reception method, comprising: performing a demodulation operation for demodulating; and performing a decoding operation for decoding data demodulated using the decoding method according to claim 15. 21. The demodulation operation according to claim 2, wherein demodulation corresponding to QPSK type modulation is applied.
0. The receiving method according to 0. 22. The demodulation operation according to claim 2, wherein demodulation corresponding to OFDM type modulation is applied.
0. The receiving method according to 0. 23. A decoding device for decoding a received digital signal that has been digitally watermarked with a known digital watermark, for decoding at least a part of the digital signal with a parameter-adjustable iterative decoder. A digital watermark extracting means for extracting a digital watermark from the decoded signal; a comparing means for comparing the extracted digital watermark with a known digital watermark; if necessary, according to the result of the comparison. And a correcting unit for correcting at least one decoding parameter. 24. The correction means, comprising: means for calculating the number of repetitions to determine the number of repetitions to be applied as a decoding parameter according to the result of the comparison; and applying the previously determined number of repetitions during decoding. 24. The decoding apparatus according to claim 23, further comprising: means for adjusting decoding for this purpose. 25. The decoding device according to claim 1, wherein the decoding unit is a partial decoding unit for decoding data by an iterative decoder by applying repetition or half-iteration. Outputting a digital signal with a mark, the correcting means includes quality testing means for testing whether the quality of the extracted digital watermark is satisfactory,
Performing additional iterations or half-iterations during decoding so that the partial decoding means finally provides the optimal number of iterations or half-iterations to apply as parameters for decoding, unless the quality is satisfactory. The decoding device according to claim 23, wherein: 26. The decoding apparatus according to claim 23, wherein the iterative decoder is a turbo decoder. 27. The decoding device according to claim 23, wherein the digital signal is an image signal. 28. A receiving apparatus for receiving a digital signal, comprising: receiving means for receiving a modulated symbol; and demodulating means for demodulating the received modulated symbol and outputting demodulated data. A decoding device for decoding the demodulated data by the decoding device according to claim 23. 29. The demodulation means according to claim 2, wherein demodulation corresponding to QPSK type modulation is applied.
9. The receiving device according to 8. 30. The demodulation means according to claim 2, wherein demodulation corresponding to OFDM type modulation is applied.
9. The receiving device according to 8. 31. A digital signal processing device comprising means adapted to implement the transmission method according to claim 1. 32. A digital signal processing device comprising means adapted to implement the decoding method according to claim 15. 33. A digital signal processing device comprising the transmission device according to claim 8. 34. A digital signal processing device comprising the decoding device according to claim 23. 35. A telecommunications network comprising means adapted to implement the transmission method according to claim 1. 36. A telecommunications network comprising means adapted to implement the receiving method according to claim 20. 37. A telecommunications network comprising the transmitting device according to claim 8. 38. A telecommunications network comprising the information receiving device according to claim 28. 39. A mobile station in a telecommunications network comprising means adapted to implement the transmission method according to claim 1. 40. A mobile station in a telecommunications network comprising means adapted to implement the receiving method according to claim 20. 41. A mobile station in a telecommunications network, comprising the transmitting device according to claim 8. 42. A mobile station in a telecommunications network, comprising a receiving device according to claim 28. 43. A base station in a telecommunications network comprising means adapted to implement the transmission method according to claim 1. 44. A base station in a telecommunications network comprising means adapted to implement the receiving method according to claim 20. 45. A base station in a telecommunications network, comprising a transmitting device according to claim 8. 46. A base station in a telecommunications network, comprising the receiving device according to claim 28. 47. An information storage medium that can be read by a computer or a microprocessor that stores instructions of a computer program and implements the transmission method according to claim 1. 48. The information storage medium according to claim 47, wherein the information storage medium is partially or wholly removable. 49. An information storage medium that can be read by a computer or a microprocessor that stores instructions of a computer program and implements the decoding method according to claim 15. 50. The information storage medium according to claim 49, wherein the information storage medium is partially or wholly removable. 51. A computer program product comprising a sequence of instructions for implementing the transmission method according to claim 1. 52. A computer program product comprising a sequence of instructions for implementing the decoding method according to claim 15.